© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 57668
    Prev
    Next
    INDEX
    Explanations

    air defence radar

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron specifically fires on tokens describing military air operations, especially references to scrambling, intercepting, or tracking combat aircraft and fighter jets.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enedig
    -0.72
     blossoms
    -0.69
    bnf
    -0.69
     cubierto
    -0.68
     microns
    -0.66
     centimeters
    -0.66
    촬
    -0.66
     outcome
    -0.66
     desembar
    -0.66
     Stefanie
    -0.66
    POSITIVE LOGITS
     NOR
    1.62
    NOR
    1.60
     radar
    1.48
     Radar
    1.30
    radar
    1.22
     defence
    1.16
    Radar
    1.13
     Fighter
    1.07
     interceptor
    1.04
    defence
    1.03
    Activations Density 0.036%

    No Known Activations